Beruflich Dokumente
Kultur Dokumente
IN THE PHILIPPINES:
FROM QUANTITY TO QUALITY AND BEYOND
1. Introduction
2. Inputs, outputs and the quality of education
2.1. International comparisons: education and productivity
2.2. Basic indicators of education in Philippines: efficiency and effectiveness
2.2.1. Primary and secondary education
2.2.2. Higher education
2.2.3. Inputs, outputs and efficiency: the regional dimension
3. Education and labor market outcomes
4. Regional shocks and workers education
4.1. Persistence of geographical differences in unemployment rates by skill level.
4.2. How do workers with different skill levels adjust to shocks?
5. Equity in the access to education
5.1. Basic expenditure indicators
5.2. Educational attainment and enrolment
5.3. Equity in the access to higher education
5.3.1. Education expenditure and access
5.3.2. Reasons for not being enrolled in education
5.3.3. Benefit incidence analysis of public expenditure in education
6. The return to education in Philippines
6.1. Previous studies on the return to education in Philippines
6.2. The return to education in Philippines using the APIS 2002.
7. Conclusions
References
Technical note I.
1. INTRODUCTION
Education is a basic factor in economic development. At the microeconomic level
education has an important role in social mobility, equity, public health, better
opportunities for employment (lower unemployment and higher wages), etc. In the case
of the Philippines the previous Poverty Assessment (World Bank 2001) showed clearly
that the educational attainment of the head of the household was the single most
important contributor to the observed variation in household welfare.
However it is also well known that the workers of Philippines have one of the highest
levels of education of Asia, specially when considering its level of development.
Probably Philippines is the most typical case of what is called the education puzzle.
Therefore the level of poverty of the Philippines is difficult to be explained by the level
of education of their workers.
The objective of this report is to analyze and propose recommendations for the situation
of the educational system of Philippines, specially with respect to the sector of tertiary
education, digging into the contribution of education to economic growth in the
Philippines as well as the factors that can explain why education is not translated into
development.
Our methodological approach is to deal analytically with all the issues, trying to find the
most recent evidence on the diagnostic and trends. The objective is to obtain the
relevant information to complement the last Philippines Poverty Assessment and cover
the period of time since it was written 1.
The organization of the report is guided by the relevant issues and not by the level of
education. Therefore, instead of dividing the sections in primary, secondary and higher
education we segment the report by issues and deal with the relevance of them for each
level of education inside each section. The plan of the report is the following. Section 1
contains this introduction. Section 2 presents an analysis of the efficiency and
effectiveness of education in terms of inputs, outputs and quality. Section 3 considers
the efficiency of education in terms of outcomes. Section 4 discusses the likelihood of a
preventive effect of education against negative shocks. Section 5 analyzes the equity of
the education system. Section 6 presents an analysis of the rate of return of education in
Philippines and its recent evolution. Finally section 7 summarizes the basic diagnostic
and recommendations offered by the report.
1
In terms of statistical information the previous Poverty Assessment stops in 1997-98.
2. INPUTS, OUTPUTS AND THE QUALITY OF EDUCATION
In this section we review the basic indicators of inputs, outputs and quality of the
education system of Philippines. First of all we present some international comparison
of inputs, outputs and quality of education for countries in South East and the Pacific
area of Asia. Secondly, we discuss inputs, outputs and quality of education in
Philippines by level of education. Thirdly we present a preliminary discussion of the
regional dimension of education in Philippines.
The level of education of population of the Philippines is much higher than the one that
corresponds to its level of development. Not only that but would score high even in
comparison with many developed countries. Table 2.1 shows has very high gross
enrolment rates in secondary and tertiary education. It is interesting to notice that the
enrolment rate in tertiary education is very high in Philippines.
Figure 2.1 shows that the enrolment in secondary education in Philippines is clearly
over the regression line of enrolment on GNI per capita. This implies that Philippines
scores much higher than countries in its geographical area with a similar level of
development.
Figure 2.1. Enrolment in secondary education and GNI per capita.
100
KOR
Gross enrollment secondary education
PHI THA
80
MON
VIETCHI MAL
60
INDO
LAO
40
CAM
20
KOR
80
Gross enrollment tertiary education
40 60
THA
MON
PHI
MAL
20
INDO
CHI
VIET
LAO
CAM
0
The proportion of public expenditure on education over total public expenditure is also
high reaching 20.6% in 1999. This said the recent evolution of the proportion of public
expenditure in education over GDP shows signs of stagnation and, even worse, a
decreasing pattern. Since the data of UNESCO do not cover the most recent period we
use the data of the ADB. Notice that the public expenditure in education in the ADB
includes only the central government and, therefore, it is not totally comparable to the
UNESCO data. Figure 1.3 shows the decreasing pattern of central government
expenditure in education over GDP in the case of Philippines, Thailand and Indonesia.
By contrast Malaysia is increasing the public effort in financing education.
Figure 2.3. Recent evolution of public expenditure on education over GDP.
9%
8%
7%
6%
5% P hi l i ppi nes
T hai l and
M al ays i a
4% I ndonesi a
3%
2%
1%
0%
Therefore in recent years the differences have been reduced by the improvement of
other countries in the region and by a certain stagnation in the educational sector of
Philippines. The latest data compile by the Asian Development Bank shows several
important trends:
a. Countries like Malaysia and Thailand are catching up very quickly in terms of
higher education enrolment. In fact in 1996 only Korea had a higher gross
enrolment in tertiary education. By 2001 Mongolia, and specially, Thailand,
have overtaken Philippines in gross enrolment in tertiary education. See also
table 2 at the end of the document.
b. Higher education in the Philippines stopped the convergence process towards
developed economies by the middle of the 90s.
At first it seems difficult to make compatible the high level of educational achievement
of Philippines workers with their low level of income per capita and productivity. Table
1.2 shows the productivity of several countries of Asia calculated as gross domestic
product divided by number of employed persons. The table shows that after the crises
some countries like Singapore or Korea have returned slowly back to the level of
productivity previous to the crisis. However in the case of Philippines there is at most a
very slow recovery path. In addition the comparison of the levels shows Philippines
lacking behind other Asian countries.
Obviously the previous table is a crude and simplistic approach to productivity but it
give us a preliminary indication. Cororaton (2002) finds that the contribution of labor
quality to total factor productivity in the period 1967-72 was 2.11%, while in the period
1991-93 was only 0.16% rising to 0.52% in 1998-2000 (see figure 2.4).
1.5
0.5
0
1967-72 1973-82 1983-85 1986-90 1991-93 1994-97 1998-2000
The reason for the decline in the contribution of labor quality to total factor productivity
are diverse and complex. However there are at least two sources of problems:
a. Low and deteriorating quality of education.
b. Migration of highly qualified workers.
Therefore there are several reason why the high educational achievement of Philippines
laborers has not been translated into high productivity and a large impact on total factor
productivity being the quality of the educational system in general and, in particular, the
university sector, one of the most prominent.
Table 2.5 present several indicators of the quality of the educational services in the
Philippines. Table 2.5 shows that the Philippines have the second highest ratio of pupils
by teacher in primary education (only below Cambodia) and the highest ratio of pupils
by teacher in secondary among the Asian countries considered in table 4. More
important that this fact we will see in next section that both ratios are increasing in
Philippines while in other countries like Korea, Indonesia, China, Malaysia and Vietnam
these ratios are decreasing.
In addition the proportion of students of science and technology is low and the
percentage of university students who graduate is also low compared with other
countries.
Table 2.5. Quality indicators I.
Pupils per teacher (2001) Tertiary education (1996)
Primary Secondary S&T students % graduates
China 19.6 18.9 na 35
Korea 32.1 21.0 32.1 38
Mongolia 31.8 21.9 24 na
Indonesia 20.9 13.6 39.2 27
Malaysia 19.6 17.9 na 32
Philippines 35.4 38.3 13.7 28
Thailand 19.1 22.3 18 18
Viet Nam 26.3 26.9 na na
Cambodia 56.3 21.6 13.2 na
LAO 29.9 24.1 na na
Myarmar 32.6 31.2 55.7 30
Source: UNESCO (several years).
2
In the Philippines, TIMSS 2003 is jointly implemented by the Department of Science and Technology
(DOST), through the Science Education Institute (SEI), and the Department of Education (DepEd).
stands at the last position in the ranking of countries of Asia. In fact Philippines is
ranked in the position previous to the last among all the countries that participated in
this international survey (38). Only Morocco and South Africa were below, although
only the second country had a score significantly smaller than the one of the students of
Philippines. This is obviously a bad sign of the quality of the educational system.
Math Science
Singapore 604 568
Korea 587 549
China 585 569
Hong Kong 582 530
Malaysia 519 492
International average 487 488
Thailand 466 482
Indonesia 403 435
Philippines 345 345
Source: TIMSS 1999.
There are some other surprising results that locate Philippines in an extreme with
respect to the performance of the students. Table 2.8 shows that in general boy
outperform girls, less in mathematics than in sciences 3. However in the case of the
Philippines girls outperform boys by a huge difference. Notice that if the average
advantage of girls respect to boys in maths is 1.25 in Philippines this differential score
reaches 15 points. In the case of science the difference is even more evident. The
average of the countries included in table 6 is 11.1 in favor of boys. However
Philippines is the only country were girls outperform boys in science and the difference
is again huge.
3
A positive sign implies that girls have a higher score than boys. A negative sign implies the opposite.
Some other indicators also point towards problems with the quality of higher education
in Philippines:
a. The academic background of the faculty (2001). The majority of the faculty
(58%) has only a baccalaureate degree. Only 8% of the professor have a Ph.D. It
seems that the low qualification of Boxthe1 faculty may be related with the upgrade
Structure of the education system in Philippines
of low level institutions to the university sector.
b.1. The
Compulsory education.
average passing rate of the professional boar examination (PBE) has been
Age of entry: 6
Ageover
of many
exit: 12years, around the 40%4. This implies that 60% of the graduates will
not bea.ablePrimary
to practice the profession for which they have been preparing for.
Length of the program in years: 4
Age: from 6 to 10.
2.2. Basicb.indicators
Intermediateof education in Philippines: efficiency and effectiveness.
Length of the program: 2
In this section wefrom
Age: review 10 the recent evolution of the basic indicator of inputs, outputs
to 12.
and the2. Secondary
quality of education
education in Philippines separating basic education from higher
Length of the program in years: 4
education.
Age: from 12 o 16.
Secondary education usually lasts for four years. Compulsory subjects include
English, Filipino, Science, Social Studies, Mathematics, Practical Arts, Youth
2.2.1.Development
Primary andTraining
secondary andeducation.
Citizens Army Training. The cycle culminates in the
examinations for the High School Diploma. The National Secondary Aptitude
Test is taken at this time. It is a prerequisite for university admission
The number
3. Higher of indicators
education. on the efficiency of the different levels of education in the
Philippines is a. overwhelming.
Post-secondaryThe studies (technical/vocational)
participation of Philippines in the project EFA
Technical or vocational courses which last between three months and
(Education For All)
three has lead
years havetoa skills
very important
proficiencyinfluence
which are inmostly
the number of indicator
terminal in nature.of
efficiency forThey the are Certificate,
education Diploma
sector. Tableand2.9Associate
contains programs.
some of the most important
b. University studies.
indices. The main characteristics
i. Bachelors aredegree
the following:
a. In the elementary A Bachelor's
level publicDegree is generally
schools are theconferred after four
overwhelming years'of
majority
study. The minimum number of credits required for four-year
institutions of Bachelor's
the sector. Degrees
Private schools
ranges represent
from 120only the 10.9%
to 190. In someof the total
fields,
such as Business, Teacher Education,
number of schools. However private elementary schools are increasing at a Engineering and
Agriculture, one semester's work experience is required.
much faster rate than public elementary
ii. University second stage: schools.
Certificate, diploma and master
b. Opposite to whatdegree. happen in the elementary grade, the proportion of secondary
Certificates and Diplomas are conferred on completion of one
private school oris two high,years
reaching 41.3%
of study of thethe
beyond total secondaryDegree.
Bachelor's schoolsToinbethe
course 2002-03. admitted to the Master's Degree, students must have a general
average of at least 85 or B or 2 in the undergraduate course.
iii. University third stage: Ph. D.
To be admitted to a Doctorate program, students must have an
average of at least 1.75 in the Master's Degree. The PhD
requires a further two to three years' study (minimum)
following upon the Master's degree and a dissertation.
4 iv. Teachers
For instance in 2000 the passing rate waseducation:
37.5%. pre-primary, primary and secondary
(bachelor). Tertiary: master degree.
v. Non-formal higher education.
Two- to three-year programs are offered to train highly-skilled
technicians, office staff, health personnel. Candidates are
awarded Diplomas, Certificates or Certificates of Proficiency.
Box 2
Legal set-up and organization of the education sector
1. Basic laws.
a. Education Act of 1982.
b. Republic Act 7722 of 1994. Creation of the Commission of Higher
Education (CHE).
c. Republic Act 7796 of 1994. Creation of the Technical Education and
Skills Development Authority (TESDA).
d. Republic Act 8292 of 1997. Higher education modernization act.
e. Republic Act 9155 of 2001. Governance of Basic education act. It
provides the overall framework for (i) school head empowerment by
strengthening their leadership roles and (ii) school-based
management within the context of transparency and local
accountability. The goal of basic education is to provide the school
age population and young adults with skills, knowledge, and values
to become caring, self-reliant, productive and patriotic citizens.
2. Legal bodies governing education
a. Department of Education (DepEd). Pre-primary, primary, secondary
and non-formal education.
b. Commission of Higher Education (CHE). Higher education.
c. Technical Education and Skills Development Authority (TESDA). In
charge of post-secondary, middle level manpower training and
vocational education.
3. Structure of the Department of Education. The Department operates with
four Undersecretaries in the areas of: (1) Programs and Projects; (2)
Regional Operations; (3) Finance and Administration; and (4) Legal Affairs;
four Assistant Secretaries in the areas of: (1) Programs and Projects; (2)
Planning and Development; (3) Budget and Financial Affairs; and (4) Legal
Affairs. Three staff bureaus provide assistance to formulate policy and
standards: Bureau of Elementary Education (BEE), cont.
c. In recent year the growth of private high schools has been similar to the growth
of the number of public high schools
d. However as a consequence of the Asian crisis many student switched from
private high schools to secondary high schools. For this reason enrolment in
public high schools is increasing fast while enrolment in private high schools is
decreasing.
Table 2.9. Schools, enrolment and teachers: elementary and secondary education.
Elementary Secondary
1997-98 2002-03 Crec. 1997-98 2002-03 Crec.
SCHOOLS 38,395 41,288 7.53 6,690 7,890 17.94
Public 35,272 36,759 4.22 3,956 4,629 17.01
Private 3,123 4,529 45.02 2,734 3,261 19.28
ENROLMENT 12,225,038 12,979,628 6.17 5,022,830 6,077,851 21.00
Public 11,295,982 12,050,450 6.68 3,616,612 4,793,511 32.54
Private 929,056 929,178 0.01 1,406,218 1,284,340 -8.67
TEACHERS 354,063 337,082 -4.80 144,662 119,235 -17.58
Public 324,039 337,082 4.03 105,240 119,235 13.30
Private 30,024 - 39,422 -
Source: Department of Education of Philippines.
Table 2.10 presents the evolution over time of several performance indicator of the
Philippines elementary and secondary system. The cohort survival rate is the proportion
of enrolees at the beginning grade or year level who reach the final grade or year level
at the end of the required number of years of study 5. We can see in table 2.10 that
around two thirds of students complete elementary school in the expected time period.
The cohort survival is even higher in secondary education. The completion rate captures
a similar indicator. It is the percentage of first year entrants in a level of education who
complete the level in accordance with the required number of years of study. The
proportions are similar to the cohort survival rates. The drop-out rate is the proportion
of students who leave school during the year as well as those who do not return to
school the following year to the total number of students enrolled during the previous
school year. The trend of the drop-out rate is increasing over these years which is
worrisome.
Table 2.10. Evolution of the basic performance indicator for elementary and
secondary education.
5
The particular version presented in the table is from UNESCOSs EFA (Education For All) program.
1997-98 1998-99 1999-00 2000-01 2001-02 2002-03
Cohort survival rate (EFA formula)
Elementary 64.96% 64.09% 69.48% 67.21% 67.13% -
Secondary - 70.31% 69.50% - - -
Completion rate
Elementary 67.67% 68.99% 68.38% 66.13% 66.33% -
Secondary (based on First year) 69.09% 69.98% 69.89% 70.62% 71.01% -
Drop out rate
Elementary 7.39% 7.57% 7.72% 9.03% - -
Secondary 9.93% 9.09% 9.55% 10.63% - -
Achievement rate
Elementary: NEAT 50.78% 50.08% 49.19% 51.73% - 40*
Secondary: NSAT 48.66% 46.12% 54.34% 53.39% - 28.5*
Pupil-teacher ratio
Elementary 34 35 35 36 36 36
Secondary 34 35 35 36 39 40
No schools
Barangays without public elementary
school 4,231 4,819 4,710 4,569
Municipalities without a high school 26 13 5 3 6 6
Source: Authors compilation from the DepEd. *NEAT and NSAT were not administrated in 2001-02. In
2002-03 there were two diagnostic tests: in grade IV for elementary and in 1st year for secondary.
The achievement rate measures the performance of student in regular tests. The NEAT
(National Elementary Assessment Test) is the national examination which aims to
measure learning outcomes in the elementary level in response to the need tof
enhancing quality of education as recommended by the Congressional Commission in
Education. It is designed to assess abilitites and skill of grade VI students in all public
and private schools. The NSAT (National Secondary Assessment Test) is a national
examination to assess the skills of fourth year high school students. There are no clear
trends in terms of the performance of the students in this tests 6. Finally, as we
mentioned before, the ratio of students to teacher is worsening over time, specially in
secondary education.
6
With those punctuations the percentage of passers is around 75% among the elementary students and
94% in secondary.
GER in ECCD 9.86% 9.60% 10.14%
% of Gr.1 w/ ECD Exp. 51.95% 51.31% 52.68%
App. Intake Rate (AIR) 125.52% 131.26% 119.50%
Net Intake Rate (NIR) 43.59% 41.90% 45.38%
Gross Enro Ratio (GER) 100.41% 101.17% 99.61% 65.66% 62.96% 68.41%
Net Enro Ratio (NER) 83.30% 82.58% 84.04% 45.56% 41.76% 49.44%
CSR (Grade VI / Year IV) 69.47% 65.49% 73.90% 63.88% 56.71% 71.22%
Completion Rate 66.94% 62.77% 71.56% 58.62% 51.11% 66.38%
Coefficient of Efficiency 81.03% 77.75% 84.47% 70.69% 64.12% 77.04%
Years Input Per Graduate 7.41 7.72 7.10 5.66 6.24 5.19
Graduation Rate 95.89% 95.17% 96.58% 90.62% 88.41% 92.58%
Ave. Promotion Rate 93.42% 92.32% 94.58% 83.82% 78.49% 88.97%
Ave. Repetition Rate 2.25% 2.91% 1.56% 2.81% 4.35% 1.32%
Ave. School Leaver Rate 7.45% 8.60% 6.21% 13.91% 17.14% 10.70%
Transition Rate 97.58% 96.71% 98.48% 92.95% 96.80% 89.30%
Ave. Failure Rate 5.24% 5.99% 4.44% 9.60% 12.59% 6.72%
Retention Rate 93.82% 92.46% 95.27% 88.18% 84.39% 91.91%
Ave. Dropout Rate 1.34% 1.69% 0.98% 6.58% 8.92% 4.31%
With respect to the quality of public and private school there are not very many
indicator since most of the EFA indicator distinguish between men and women but not
by the ownership of the schools. Indicator 11 presents the calculation of the
student/teacher ratio separating public and private schools. In elementary schools the
ratio of public schools is 35.7 while in private schools is 30.1. In high schools the ratio
is 35.9 for public schools and 33.6 for private schools8. Therefore it seems that in terms
of the student/teacher ratio private schools are better equipped to produce high quality
education than public schools.
Another piece of information that may be important in rating public versus private
schools is the level of satisfaction of the users. In a recent report 9 the World Bank asked
7
Latter we will show that the return to education is also much higher for women than for men.
8
However we should notice that there is a very high variability across regions.
9
World Bank (2001), Filipino report card on pro-poor services.
Filipino10 families for their level of satisfaction with public and private schools. Overall
the level of satisfaction with public and private schools was very similar even thought
rating were higher for private schools in the quality items and for public schools in the
costs items. The present rating of public schools was 1.49 (past rating 1.50) while for
private schools it was 1.51 (past rating 1.71).
The highest satisfaction with public schools was associated with its convenient location,
consequence of the longstanding policy one-barangay, one public school. The rating
of public schools was low in class sizes, textbooks and facilities. Class sizes are larger
not because of a simple shortage of teacher but also because of a poor policy of teacher
deployment caused by the restrictive regulation on deployment of teacher in Philippines.
In addition real students to teacher ratios in public schools are higher than the number
shown by aggregate statistics due to many teachers doing clerical or administrative
functions. Real ratios are close to 45.
In private schools the highest degree of satisfaction corresponds to teachers attendance
to schools, and availability of books. The lowest satisfaction is associated with the
tuition charged by private schools.
The Filipino Record Card study provides also estimates for the cost of public and
private schools. Table 2.12 shows these findings.
Table 2.12. Annual per student cost for elementary education (in PhP)
Fees Textbooks Uniforms Transport Mean
Public 445 38 522 1,017 2,023
Urban 666 69 612 1,410 2,767
Rural 230 9 438 651 1,329
Bottom 30% 204 7 341 419 917
Middle 30% 325 17 405 794 1,200
Top 30% 757 84 785 1,706 2,065
Private 20,658
Source: Filipino report card on pro-poor services (2001)
Summarizing we can say that household would prefer to send their children to private
schools if they did not have any financial constrain. A simple indication of this is that
between 14% and 27% of households currently sending their children to a public school
rated public schools better than private school. However from 63% to 93% of
households sending their children to private schools rate them better than public
schools.
10
Ratings are calculated as the average of a 5-points Lickert scale: 2 Very satisfied; 1 somewhat satisfied;
0 undecided; -1 somewhat dissatisfied; -2 very dissatisfied.
2.2.2. Higher education.
The higher education sector of Philippines is one of the most interesting cases of higher
education in the world. With 1,479 institutions of higher education Philippines ranks
second in the absolute number of HEIs. Opposite to primary and secondary education in
the higher education sector the large majority of the centers are private institutions. By
1965 private HEI were 94% of the total number of HEIs. The proportion decreased until
the end of the 80s were it reached around 72% and increased again during the 90s. The
proportion of private institutions in the course 2002-03 was 88,2%. Figure 2.5 shows
that most of the recovery of the private HEIs during the 90s comes from the
contribution of the non-sectarian institutions.
1,200
1,000
800
Sectar i an
600 Non-sectar i an
P ubl ic
400
200
1990-1991 1991-1992 1992-1993 1993-1994 1994-1995 1995-1996 1996-1997 1997-1998 1998-1999 1999-2000 2000-2001 2001-2002 2002-03
Source: CHED.
Usually HEIs are divided in public, private sectarian and private non-sectarian. The
sectarian institutions are mostly religious and non-profit. They represent the high end of
quality in the private sector. However most of the private HEI are non-sectarian
(75,1%), for profit schools with large classes, scarcely selective admission processes
and low tuition. The public institutions can be divided in SUCs (state and university
colleges), CHIs (CHED supervised institutions) and LUCs (local universities/colleges).
Public institutions charge low tuition fees , selective process of admission and very high
unit costs. Table 2.13 presents the recent evolution of the number of HEI classified by
groups.
Table 2.13. Higher education institutions in Philippines.
Sector/Institutional Type 1998-991999-002000-012001-022002-03
Table 2.14 presents the distribution of students by disciplines. The largest proportion
corresponds to business administration and related disciplines (25.9%) followed by
education and teaching (17.8%) and engineering and technology (15.3%). Mathematics
and computer sciences is the choice for 10.6% of the students. The trends are also
interesting. In recent years teachers training has reduced its proportion in the number of
students while business administration has stabilized and technology, mathematics and
computer science are growing.
Table 2.15 presents the recent evolution of graduates from each discipline. Half of the
graduates have majored in business administration or education.
Table 2.16 present several indicators of the intensity of graduation over number of
students by disciplines. The highest proportion of graduates over students is found in
business administration, mass communication, medical disciplines and social and
behavioural sciences. Architecture and law have the lowest ratios.
Table 2.16. Basic proportions and ratios by disciplines (2001-2002).
Discipline Group % students % graduates ratio grad/stud
Agricultural, Forestry, Fisheries, Vet
Med. 3.85 3.62 0.14
Architectural and Town Planning 1.02 0.70 0.10
Business Admin. and Related 25.97 29.33 0.17
Education and Teacher Training 17.82 19.61 0.16
Engineering and Technology 15.30 12.42 0.12
Fine and Applied Arts 0.36 0.36 0.15
General 1.77 1.51 0.13
Home Economics 0.26 0.26 0.15
Humanities 1.20 1.16 0.14
Law and Jurisprudence 0.80 0.60 0.11
Mass Communication and
Documentation 1.24 1.41 0.17
Mathematics and Computer Science 10.63 9.04 0.13
Medical and Allied 6.65 7.51 0.17
Natural Science 1.23 1.32 0.16
Religion and Theology 0.32 0.29 0.13
Service Trades 0.63 0.65 0.15
Social and Behavioral Science 3.25 3.68 0.17
Trade, Craft and Industrial 0.19 0.20 0.15
Other Disciplines 7.51 6.32 0.12
Source: Authors calculations using CHED data.
The ratios in the previous table are affected by the differential growth in the number of
student in each discipline and, therefore, it is difficult to interpret outside a steady state.
Table 2.17 presents aggregated indicators not subject to this problem.
Indicator
Academic Year
Gross Enrollment Ratio
Gross Survival Rate Graduation Rate
/Participation Rate
As we shown before the gross enrolment in higher education has continue to grow
during recent years reaching 22% of the relevant group age. However the survival rates
and, specially, the graduation rates are very disappointing. Only 46.5% of the student
graduate. This problem is compounded by the fact that only around 45% of those that
graduate are able to pass the Professional Board Examinations. This is a national test
which covers most of the fields of study. Among the large discipline the only exception
is business and commerce. We should also notice that not all graduates take the exam
(the weakest graduates do not take the exam) which means that the effectiveness of the
higher education system of Philippines is even worse than what nominal rates of failure
in the exam would suggest.
Figure 2.6 shows the recent evolution of the percentage of graduates passing the exam.
Even thought in recent years the proportion of graduates failing the exam has been
slowly decreasing there is a suspicion that the difficulty of the exam is going down.
60%
50%
40%
30%
20%
10%
0%
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
The previous sections present a unifying view of inputs, outputs and efficiency in the
educational system of Philippines. However there are large regional differences in
inputs and outputs. In this section we only scratch the surface of this problem. Table
2.19 presents the basic data on inputs by regions. As expected net enrolment is
negatively correlated with the level of development of each region. This effect is more
evident in secondary education than in primary education.
The relationship between inputs ratios (pupil/teacher ratio, pupil/seats ratio, etc) and
poverty is more complex. In the less developed regions those ratios are high but they are
also quite high in the most developed areas of the country (like NCR), at least in
elementary school. In high school the correlation between average poverty and inputs
ratios is again quite clear. Figure 2.19 present a map with the ratio students/teacher in
secondary education by region. We should again emphasize that these are nominal ratios
in the sense that the denominator included all the teachers. Since many of them are
doing clerical jobs the real ratios are suspected to be higher than those in the map.
Table 2.20 contains the outcomes of the primary and secondary education system by
regions. As expected dropout rates, survival rates and scores are positively correlated
with the level of development of the regions.
Table 2.19: Basic indicators by regions. Elementary and secondary. SY 2002-2003.
Nationally Net Pupil Pupil Pupil
REGION Enrolment Funded Enrolment. Teacher Instructional Seats
Teachers Ratio Ratio Room Ratio Ratio
2000
ELEMENTARY
SECONDARY
Legend
- 24.99
25.00 - 29.99
30.00 - 34.99
35.00 - 39.99
40.00 - 44.99
45.00 - 49.99
CAR 50.00 +
STR = 34.95 No Teachers
Region I No Data
STR = 36.78 Region II
STR = 41.06
Region V
NCR
STR = 38.49
STR = 35.88
Region IV-B
STR = 40.01
Region VIII
STR = 40.84
Region VI
STR = 37.29
Region IX CARAGA
STR = 40.1 STR = 44.25
Region XI
STR = 42.15
11
The annual unemployment rate is calculated as a simple average of the quarterly unemployment rates.
The definition for unemployed is the one used by the NSO. Using the ILO definition provides very
similar conclusions.
Figure 3.1. Unemployment rate by level of education.
1996 1998 2000 20021996 1998 2000 20021996 1998 2000 2002
year
Graphs by grade
1996 1998 2000 20021996 1998 2000 20021996 1998 2000 2002
year
Graphs by grade
Using the Tracer Study12 of the CHED (1999) we can understand a little better the
reasons for the unemployment of the college educated active population.
In terms of the quarterly volatility of the indicators we can see in figure 3.4 that again
the undergraduates of high school and college are the ones with the highest degree of
between-quarter volatility. The rest of the levels of education present a very narrow
difference across quarters.
12
The Trace study is a continuous survey on the situation of graduates from higher education institutions.
Although the representativeness of the sample is not clear (students can fill the questionnaires using
Internet with all the sampling problems that this strategy generates). Notice also that this study refers
basically to recent college graduates and undergraduates.
Therefore the first type of mismatch is the high level of unemployment of college and
higher education workers in comparison with lower levels of education. The second
type of mismatch is the increasing lack of relationship between the field of study and
the field of work. The Tracker Survey confirms that the percentage of graduates
working in jobs requiring the field studies they have has decreased over time. A third
type of mismatch is the question of overqualification. There has been an artificial
demand for higher education since there is an increasing nominal demand for college
graduates13 by employers. This effect would correspond to the credentialist view of
education: since there is imperfect information about the ability of workers employer
use credentials as proxy for productivity. However, this does not guarantees that
workers will work in jobs for which their level of education is required. Table 3.2
shows the distribution of the primary occupation of college educated workers in the
October round of the 2002 LFS. We can see that the proportion of clerks, sales workers,
pan operators and elementary occupation is very high. In fact the level of
overqualification reaches the 55.9% (assuming that the technicians and associated
professionals need to be college educated).
Finally another form of mismatch is the existence of many college educated who do not
participate in the labor force. This has been an argument in several recent report.
However looking at the participation rates of college educated people we can see that
college graduates, as expected, have a higher labor force participation rate than any
other educational level. Nevertheless if we aggregate graduates and undergraduates the
picture is quite different since college undergraduates have very low rates of labor force
participation rates and, as in the case of high school undergraduates, very volatile
(figures 2.3 and 2.4).
.8
.7
labor force participation
.6
.5
1996 1998 2000 20021996 1998 2000 20021996 1998 2000 2002
year
Graphs by grade
1996 1998 2000 20021996 1998 2000 20021996 1998 2000 2002
year
Graphs by grade
4. REGIONAL SHOCKS AND WORKERS EDUCATION.
The analysis splits the population of workers into three groups: workers with education up
to primary graduates; workers with high school education and finally the third group are
workers with college education.
There is evidence that the pattern of unemployment across groups has persisted for many
years. Scatter plots of the unemployment rates in first quarter of 1993 and last quarter of
2002 for the 14 regions in Philippines reveal a remarkable correlation between the
provinces that have higher unemployment rates in the 2002 and those that had higher
unemployment rates in the 1993 (Figure 3.1 Figure 3.4). The degree of persistence of
geographical differences in unemployment varies depending on the labor force
participants' skill levels. Workers with primary and high school education display higher
unemployment persistence then college educated workers, as can be seen in the scatter
plots of the unemployment rate in 1993 and 2002.
What is a bit surprising is that the primary education workers display lower persistence of
unemployment then those high school educated, although the difference is small. Table 3.1
reports, for each educational group, the coefficient of correlation between unemployment in
1993 and unemployment in 2002.
Table 4.1. Unemployment Persistence by Educational Group, 1993-02
When labor market experiences a negative employment shock, workers in a given region
can basically react in three ways: they can remain unemployed, drop out of the labor force
(become discouraged workers), or migrate. There are several reasons 15 to expect different
responses within groups with different level of education. In the follow up we investigate
those differences estimating a VAR system and confront them with those that usually
exist in developed countries.
The relative speed and strength of the adjustment mechanisms described above is
estimated using a panel vector autoregression (VAR) system of employment growth, the
employment rate, and labor force participation, for the 14 regions in Philippines 1993-2002.
The framework adopted is identical to that developed by Blanchard and Katz (1992). The
system is the following:
where all variables are differences between province i and the national average, in order to
focus on developments at the provincial level that are not due to nationwide developments.
ei 16
is the first difference of the logarithm of employment; lei is the logarithm of the
14
Compare Mauro (1999)
15
For example opportunity cost of not working is higher for the highly skilled; more discussion in Mauro
(1999)
16
Given that we dont have the level of unemployment we obtained this variable in the following way:
E E W E i ,t 1 E t 1 Wi ,t 1
ei ln it ln t ln it ln ln ln , where Wit and Eit are
Wt 1
Wit Wt Wt Wi ,t 1 Wt 1
the working-age population and the number of employee in state i, in time t, respectively and the variable
ratio of employment to the labor force; and lpi is the logarithm of the ratio of the labor force
to the working-age population. There is one lag for each right-hand side variable, to allow
for feedback effects from labor force participation and the employment rate to employment
growth. The system is estimated by pooling all observations, though allowing for different
province-specific constant terms in each equation, using the data for each educational
group.
As we can see in the impulse response graphs (Figures 4.5-4.7) based upon the
estimated parameters of the system above, in general a negative shock to labor demand
produces the following effects: immediately after the shock the participation rate decreases,
unemployment increases and the level of employment drops.
The NEXA describes the sources of funds for education as well as the uses of the funds.
With respect to the sources of funds NEXA uses the typology of economic transactions
in the 1993 UN System of National Accounts, adopted by the Philippines System of
National Accounts. The institutional units distinguish five resident institutional sectors:
general government, households, financial corporations, non-financial corporations and
non-profit institutions serving households (NPISH).
Table 5.4 shows the aggregate expenditure in education by items calculated from the
Family Income and Expenditure Survey. The first thing to notice in table 1.8 is the large
discrepancy between total expenditure in 1997 calculated from the FIES97 (58,2
17
Notice that the proportions by level of education do not coincide with the ones presented in table 1.2
because of differences in definitions.
billions of Pesos18) and the value of the household expenditure in education that appears
in the NEXA education accounting (94,2 billions). However we should notice that the
accounting of NEXA includes many levels of education and items of expenditure that
are not considered in the Family Income and Expenditure Survey.
The growth rate from 1997 up to 2000 was 18% in constant prices. Fees are the fastest
growing item of education expenditures. This is important since fees represent 95.3% of
the total education expenditure by families. Notice that since GDP grew 14% in
constant prices during the same period it looks as if families have increase their
education expenditure faster than the growth rate of the economy.
However the figures in table 5.4 refer to the total expenditure. Table 5.5 presents the
average family expenditure in education19. The average family expenditure in education
has growth only 5.6% during the period 1997-2000, well below the growth rate of the
economy. As before the fees are the item that grew faster.
18
This includes the sum of educational expenditure in cash and in kind.
19
We will analyze the per capita expenditure in education in the section on the incidence of public
subsidies in education.
Table 5.6 present the average household expenditure in education and the average
proportion over total expenditure for families that expend any positive amount on
education. Households are classified in function of their decile in income per capita.
Since education is a normal good we see that the average proportion of expenditure on
education increases with income per capita.
0.9
0.8
0.7
0.6
Proportion
0.5
0.4
0.3
0.2
0.1
0
I II III IV V VI/VII SEC I SEC II SEC III SEC IV
Grade
0.9
0.8
0.7
0.6
Proportion
Richest 20%
0.5 Middle 40%
Poorest 20%
0.4
0.3
0.2
0.1
0
I II III IV V VI/VII SEC I SEC II SEC III SEC IV
Grade
0.9
0.8
0.7
0.6
Proportion
Male
0.5
Female
0.4
0.3
0.2
0.1
0
I II III IV V VI/VII SEC I SEC II SEC III SEC IV
Grade
0.9
0.8
0.7
0.6
Proportion
15-19
0.5 20-29
30-39
0.4
0.3
0.2
0.1
0
I II III IV V VI/VII SEC I SEC II SEC III SEC IV
Grade
0.9
0.8
0.7
0.6
0.4
0.3
0.2
0.1
0
I II III IV V VI/VII SEC I SEC II SEC III SEC IV
Grade
Figure 5.6 shows that the gap in educational attainment between the richest and the
poorest decile in per capita income has been reduced in the latest generation (15-19 year
old) with respect to the previous one (20-29 year old), even though it continues to be
quite important. The reduction on this gap is also very significant in the difference
between educational attainment of urban versus rural youngsters as shown in figure 5.7.
0.9
0.8
0.7
0.6
15-19. Urban
Proportion
20-29. Urban
0.5
15-19. Rural
20-29. Rural
0.4
0.3
0.2
0.1
0
I II III IV V VI/VII SEC I SEC II SEC III SEC IV
Grade
0.9
0.8
0.7
0.6
15-19. Males
Proportion
20-29. Males
0.5
15-19. Females
20-29. Females
0.4
0.3
0.2
0.1
0
I II III IV V VI/VII SEC I SEC II SEC III SEC IV
Grade
0.9
0.8
0.7
0.6
Proportion
2002
0.5 1998
1993
0.4
0.3
0.2
0.1
0
I II III IV V VI/VII SEC I SEC II SEC III SEC IV
Grade
Source: Filmer and Pritchett (1993, 1998) and authors calculations using APIS 2002
Figure 5.10 considers the evolution over time of the educational attainment of the 15-19
age group. The calculations for 1993 and 1998 are taken from the educational
attainment project of the World Bank20. The improvement between 1998 and 2002 is
small but noticeable. Figure 5.11 shows the same evolution over time of educational
attainment but dividing by decile in income per capita. The change between 1993 and
1998 was very small in the gap between the educational attainment of the children of
the poorest and the riches families. In 2002 the gap has gone down very much despite
the fact that the children from poor families have still a much lower level of educational
attainment than their richest counterparts.
0.9
0.8
0.7
0.3
0.2
0.1
0
I II III IV V VI/VII SEC I SEC II SEC III SEC IV
Grade
Source: Filmer and Pritchett (1993, 1998) and authors calculations using APIS 2002
0.95
0.9
0.85
0.8
Proportion
0.75
0.7
0.65
0.6
0.55
0.5
6 7 8 9 10 11 12 13 14
Age
0.95
0.9
0.85
0.8
Proportion
Rich 20%
0.75 Middle 40%
Poorest 40%
0.7
0.65
0.6
0.55
0.5
6 7 8 9 10 11 12 13 14
Age
0.95
0.9
0.85
0.8
Proportion
Male
0.75
Female
0.7
0.65
0.6
0.55
0.5
6 7 8 9 10 11 12 13 14
Age
Figure 5.15 shows that rural children have a much lower enrolment rate than urban
children. This is observed not only for the final courses of secondary education but it
can be traced even to the first year of elementary education.
Figure 5.15. Enrolment by urban/rural location.
0.95
0.9
0.85
0.8
Proportion
Urban
0.75
Rural
0.7
0.65
0.6
0.55
0.5
6 7 8 9 10 11 12 13 14
Age
0.9
0.8
0.7
0.6
Proportion
2002
0.5 1998
1993
0.4
0.3
0.2
0.1
0
6 7 8 9 10 11 12 13 14
Age
Source: Filmer and Pritchett (1993, 1998) and authors calculations using APIS 2002
Figure 5.16 shows an amazing improvement in the enrolment rate of 6 and 7 years old
from 1993 to 2002. In only 10 years the enrolment rate of 6 years old have grown from
5% to 76%. The improvement is less noticeable, although also visible, for higher age
groups.
0.9
0.8
0.7
0.6
Richest 20%. 2002
Poorest 40%. 2002
Proportion
0.3
0.2
0.1
0
6 7 8 9 10 11 12 13 14
Age
Source: Filmer and Pritchett (1993, 1998) and authors calculations using APIS 2002
Figure 5.17 shows the enrolment rate by year and income per capita. The improvement
in the general rate of enrolment by age covered clear differences in the rates by income
per capita. For instance if in 1993 6 years old children of both, rich and poor people,
had a very lows levels of enrolment the improvement over the last decade has been
much more pronounced for children in rich families than for children in poor families.
At the other end, the highest grades of secondary education, we can observe the same
effect. The general improvement on enrolment has benefit much more children from
rich families. If we look at the difference of rate between children of rich and poor
families in 1993 we observe a smaller gap than in 2002.
5.3. Equity in the access to education.
We can look at this problem from at least from two more alternative perspectives: the
expenditure on education of the households and school attendance by income group.
The evolution of expenditure by the level of income of the households shows that
inequality in the average expenditure by decile has increased. For instance the lowest
decile spend in 2000 about 4 times more than in 1988. However the highest decile has
increase its expenditure in education by a factor of 5.8. This is particularly important if
we take into account the reduction in social expenditure during the 90s.
Tan (2002) has pointed out that even in State Colleges and Universities (SCU) less than
30% of the students come from the low deciles of the distribution.
Some theories argue that what matter is not only the average level of education but also
its distribution. The unequal distribution of education tend to have a negative impact on
income per capita. Table 5.8 shows the Gini index of education achievement. Although
the index of Philippines is lower than the other countries, with the exception of Korea,
the reduction during the decade of the 90s has been very slow.
The coefficient of variability shows very similar trends in education inequality (see
table 1.10).
In this section we analyze the questions in the APIS on the reasons why a child is not
attending school. The APIS of 1998 and 2002 include a question on the reasons why a
child in a particular household is not attending school. The answer to this question can
help to understand the problems for access to education of important groups of
population. Table 5.10 presents the proportion for each reason for all the individuals in
the surveys. The comparison of 1998 and 2002 shows that the percentage that grows the
most is the inability to pay the high cost of education. On the contrary the reasons that
fall the most are housekeeping and the lack of personal interest.
Table 5.10. Reasons for not being enrolled.
1998 2002Change
Schools are far/No school w/n brgy 1.09 1.26 0.17
No regular transportation 0.33 0.29 -0.04
High cost of education 19.25 22.7 3.45
Illness/Disability 2.80 2.75 -0.05
Housekeeping 10.27 7.56 -2.71
Employment/Looking for work 26.87 26.77 -0.10
Lack of personal interest 21.55 19.17 -2.38
Cannot cope with school work 2.50 2.23 -0.27
Finished schooling 6.91 10.05 3.14
Others 8.43 7.22 -1.21
Source: authors calculation using APIS 1998 and APIS2002
We can also analyze the effect of these reason by the theoretical level of education that
the child would be attending. For this purpose table divides the proportions by age of
the child. It is interesting to notice how the percentage of importance of high cost of
education increases up to high school and decreases a little bit for higher education.
Table 5.11 shows also that the fact of not having an educational institution close is
important for pre-primary and elementary education but it is not considered as a
problem for high school and university.
Another way of looking at the relationship between poverty and education is to look at
the benefit incidence of public expenditure in education21. From the previous indications
it seems clear that expenditure in elementary and secondary schooling at least is not
regressive. The previous PPA showed that overall public expenditure in education was
mildly progressive. However it was noticed that while expenditure in elementary
education was pro-poor, expenditure in secondary education was neutral and
expenditure in higher education was quite regressive. The data in the APIS 2002 do no
allow to make these calculations since it has eliminated the question on the type of
school from the individual questionnaire. The calculation using the APIS 1999 lead to
the same conclusions as the previous PPA: tertiary education is highly regressive. Figure
5.18 shows the cumulative number of beneficiaries of public expenditure by income per
capita decile.
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
P r o po r t i o n o f po pul a t i o n ( r a nk e d by i nc o me pe r c a pi t a )
21
See Demery (2000).
6. THE RETURN TO EDUCATION IN PHILIPPINES.
Gerochi (2002) has also analysed the issue of the return to education. Gerochi (2002)
describes the difference between successive levels of education over time using the LFS
as the source of data.
Recently Krafts (2003) has calculated the private and social rate of return of education
in Philippines using the 1998 APIS for the subset of wage and salaried workers. Table
6.3 presents the results, were there is a distinction between complete and incomplete
cycles22.
Table 6.3. Rates of return III.
Private Social
Complete cycle
- High school 10.44 7.26
- College 13.53 10.55
Incomplete cycle
- Elementary graduate 11.10 6.44
- Some high school 6.25 3.69
- High school graduate 10.16 6.33
- Some college 8.93 6.21
Source: Kraft (2003).
The rates of return on table 6.3 are lower than the ones presented above. Perhaps more
surprisingly the difference between the private rate of return of college education
(13.53) and elementary education (11.10) is low. If we were to factor in the different
probability of unemployment of both levels of education then the relative advantage of
tertiary education would be under question.
Nevertheless we have to take the estimates in table 6.3 with caution. The previous tables
show how different studies find very different estimates of the return to education. We
will consider this issue in following sections in order to clarify the differences and
provide our own estimates.
22
Krafts (2003) argues that the rate of return for elementary graduates and some elementary education is
too small for obtaining reasonable estimates.
Box 3
Data limitations on wages.
The study of the return of education requires, as a basic input, information about
wages and hours of work. It is well known that usually the LFS do not include
questions on salaries while Family Income and Expenditure Surveys do not consider
hours of work as relevant information. In the case of the statistical information of
Philippines the situation is more complex because of two factors:
- The excessive change in the questionnaires of otherwise identical statistical
operations.
- The special nature of some variables in surveys that, for most of the variables,
are publicly available.
We should notice the following points:
- By complementary information it is clear that there is information on wages
(basic pay) in the October wage of the LFS of Philippines. The files that we
have do not provide that information.
- In addition, from January 2001 onwards there should be information on wages
for all the waves. The files we have for 2001-2002 do not contain such
information.
- The other recent sources of information on wages and salaries are:
o FIES 1997
o APIS 1998
o APIS 1999
o FIES 2000
o APIS 2002
- Unfortunately the APIS 2002 do not contain information on hours worked. There are
some cases in which a panel data can be constructed. This is the case of the APIS 2002
and the October 2002 LFS. However the matching is less than perfect (see appendix
for details on the matching of these two surveys).
The literature has proposed many different estimators and corrections to calculate the
return to educations. However in this section we are not going to present an academic
discussion on the advantage and disadvantages of different methods of estimations. In
fact, and in order to establish reasonable comparison with previous results 23, we plan to
use the simple Mincerian equation for alternative samples of the individuals in the APIS
2002. We interpret the results are descriptive statistics instead of given an structural
interpretation of the coefficients. We restrict the sample to individuals between 25 years
old and 65 years old. In addition we consider only wage earners 24 since the inclusion of
self-employed is problematic for this kind of studies.
The variables included in the regression are the number of years of education (neduc),
the potential number of years of experience (calculated as age minus number of years of
education minus 6) and the square of the number of years of experience25.
Table 6.4. Return to education (wage earners between 25 and 64 years old)
All Male Female
Coeff. t-stat Coeff. t-stat Coeff. t-stat
neduc 0.1576 97.49 0.1423 72.99 0.2069 73.08
exper 0.0364 17.05 0.0400 15.41 0.0245 7.02
exper2 -0.0005 -13.02 -0.0005 -11.49 -0.0003 -4.38
Adj R2 0.33 0.28 0.48
N 21,377 14,311 7,066
Fuentes: Authors calculations based on APIS2002.
The number of years of education have been assigned using the variable that contains
the highest level of education. Since this variable distinguishes between graduates and
non graduates there are several categories that have the same number of years of
education. For instance graduates of high school and individuals that have done only
until the 4th grade of high education are assigned 10 years of education26.
The endogenous variable is the logarithm of the salary per hour. This variable is
problematic since the APIS 2002 do not contains any indicator of hours worked. For this
reason it was necessary to merge the APIS 2002 with the October round of the Labor
23
In particular with the results in the previous Philippines Poverty Assessment (2001).
24
Wage earners are defined by the variables col20_cworker. We include as such the workers for private
households, for private establishment, for government and government corporations and workers with pay
on own family operated business.
25
We should notice that in most of the regressions the product of education by experience is significantly
different from 0. In order to be able to compare with previous results we keep exactly the same
specification.
26
We discuss the return of degrees latter in this section.
Force Survey. We use the variable normal working hours from the LFS as the
denominator for the salaries in order to obtain the salary per hour.
Table 6.4 shows that the average rate of return for a year of education is 15.7%. This is
quite high for international standards. Separating males and females we find something
what many other studies have found in the past: males have a lower rate of return to
education (14.2%) than females (20.6%). Additionally the fit of the regression for
females is almost double (0.48) the fit of the regression for males. This implies that
education and experience explain a much higher degree of salaries variability in the case
of females than in the case of males.
Table 6.5 shows the same regression but restricting the sample to the household head, if
he/she is a wage earner between 25 and 64 years old. The results show a smaller return
to education (13.6%). This pattern was already present in the analysis of the return to
education in the previous Philippines Poverty Assessment. As before women (18.1%)
enjoy a better rate of return than men (13.3%).
If instead of using regression we just calculate a simple test of differences between the
group of graduates a non graduates from high school the result is the same. The
difference is not significantly different from 0 (diff= 8.9%; t=1.49).
27
The results are unchanged if we do not control for the years of experience.
However these results do not correspond to the ones obtained using the APIS98 . The
APIS 1998 sample reveals a 9.9% of advantage, in terms of wages, of elementary
graduates versus non graduates; a 13.4% in the case of higher education graduates and a
19.3% for higher education graduates.
7. CONCLUSIONS.
During the last 10 years there have been many comprehensive analysis of the situation
of education in Philippines. Five of them are particularly relevant: the Congressional
Commission on Education (1992); the Oversight Committee of the Congressional
Oversight Committee on Education (1995); the Task Force on Higher Education of the
CHED (1995); the ADB-World Bank report on Philippines Education for the 21 st
Century (1999); and the Presidential Commission on Education Reform (2000). Most of
these studies present a similar diagnostic of the problems of education in Philippines,
and in particular, higher education. The even proposed similar ideas to try to solve those
problems. However, as we will show in this section, most of the proposal have not been
developed and, in fact, many of the problems identified by these studies have worsen in
recent years. There are basically two reasons for the failure of these proposals:
a. The political economy of the education sector in Philippines is particularly
difficult. This implies that institutional inertia has a very important weight (Tan
2001).
b. There are too many proposals. Although finding a silver bullet for the problems
of education in Philippines seems an overwhelming endeavour there is need for
simple and applicable proposals that can generate political consensus, since the
institutional constrains and the effect of inertia are very important.
In the following sections we present, under separate epigraphs, a summary of the
diagnostic and recommendations derived from previous sections.
12
10
Y ear
It has been recognized for a long time that there is quite an important level of
corruption in the purchase of textbooks and the cost of constructions of classrooms. In
fact the Filipino Chambers of Commerce and Industry has shown that it is possible to
construct a classroom for PhP400,000 instead of the PhP800,000 that cost a classroom
procured through the government.
All the indicators (new TIMSS data, recent licensiture examination passing rates, etc)
show that there is not improvement in the traditional low quality of education in
Philippines, specially in secondary and higher education.
Recommendations:
Reverse the recent trends in public expenditure on basic education.
As we show previously, the private enrolment in secondary education has decrease
8,6% from 1997-98 to 2002-03, although the number of private high schools has
increase from 2,734 up to 3,261. Therefore the growth rate of private high schools in
this period (19,3%) has been higher than the growth rate of public high schools (17%)
even though the enrolment in the second type of institutions has increase 32,5%. If
students have been transfer from private high schools to public high schools for
economic reasons (Asian crisis) the voucher system should be used with more intensity
to avoid create private capacity underutilization. During the school year 2001-02 1,427
high schools participated in the voucher system. This is still only 46% of the total
private high school.
Increase the payment for the vouchers. Under the voucher system the government
pays per student PhP2,500 with a cost of around PhP700 millions. This is less than 0,7%
the total budget of the DepEd. There is no doubt that vouchers are more efficient than
the construction of new classrooms and the underutilization of private high schools. The
CPBD (2003) calculates that with PhP700 millions used for vouchers it was possible to
construct 851 classrooms (cost per classroom=PhP800,000) for 42,550 students (at the
actual rate of pupils per class). This is only 16% of the 275,000 beneficiaries of the
Education Service Contracting.
Impose a rule that force to increase operational budget in proportion to personal
budget with a target of 15% in 5 years.
Equity considerations.
Diagnostic:
Incidence analysis shows that primary and secondary education are pro-poor.
However higher education is clearly regressive.
The regressive nature of higher education hinges in two factors: students from wealthy
families can attend good high schools and get access to the best public and private
schools; and the cost of public university is twice as high as the cost of private higher
education institutions.
The low quality of secondary schools for the poor implies that their level of
knowledge is low at the end of high school and, therefore, cannot access to the good
public universities. This means they have to go to low quality-low fee private
universities if they wan to continue their education.
Due to budget limitations the number of grants for higher education has decreased
(more than 10% from 2001 to 2002).
Recommendations.
Increase the number of grants for poor people to study higher education.
Implement pro-poor systems to admission to the good public universities (where
rejections rate range between 70 and 90%). However, avoid using income as basis
for admission (income should only be used for concession of grants, conditional on
admission). It is more efficient to grant automatic admission to their preferred
higher education institution to students in the top 10% of their high school class
(ranked by the high schools themselves or in function of a national exam but ranked
by schools).
Revert the trend in the reduction of scholarships.
Regional issues
In this report we have emphasize the issue of regional differences in education inputs,
outputs and outcomes. Poor regions, in particular Mindanao, have very low levels of
participation in education and high drop-out rates. In addition regional cohort survival is
highly negatively correlated with poverty incidence. The regional dimension of poverty
is very important in educational issues.
Higher education.
The performance of higher education graduates in the job market is, in general,
disappointing. They show high unemployment rates (higher than lower levels of
education) and, even thought their underemployment rate is low, they work in jobs not
adequate for their level of education (overqualified).
There are at least to reason for the high rate of unemployment rate among higher
education graduates: it could be voluntary unemployment (higher education graduates
from wealthy families can afford to wait a long time until they find a job they want to
accept) or it could be due to low demand for university graduates because of the low
average quality of their education. The ICS of Philippines shows that firms believe
tertiary graduates do not have the necessary skills for the jobs they demand.
It is by now a common place to argue that the low level of skills and knowledge of the
average graduate from higher education has to do with the number of years of education
prior to reach that last level. In many countries students accumulate 12 years of
education before gaining access to the university. In Philippines they only need to study
for 10 years (6 years of elementary school and 4 years of high school). For this reason
some reports argue that the first two years of the university are used to rise the
knowledge of students to the level they should have had prior to enter in the university.
The number of higher education institutions continues to grow.
The proportion of graduates by disciplines is not adequate for the new era of
globalization.
Given the low enrolment rate of the poor in tertiary education there should be an
increasing amount of scholarships. However, the number of beneficiaries of the CHED
student financial assistance program went down 10,2% from 2001 (44,876) to 2002
(40,294). This trend is obviously quite regressive.
Recommendations.
Establish a moratorium in the number of new higher education institutions. The new
regulation should impose pre-accreditation to any institution that wants to become a
university. If the institutions do not get at least the minimum level of accreditation it
should not be allowed to operate.
Establish a minimum level of quality of any institution in the higher education sector.
For private institutions there should be regulation and periodical accreditation. For
public institutions the financing should change from the automatic increase to a
selective financing system in function of outputs and outcomes.
Financing formulas for allocation of public funds among public HEI should consider
not only enrolment but also outcomes. The formula should consider unit costs of the
provision of higher education but should also weight heavily outcomes (performance of
graduates in the labor market, etc.). For this formulation to work a complete and well
organized information system is needed. The TRACER system in place at the CHED to
find out about the job history of graduates is not an effective source of information since
it is subject to many types of biases (self-selection, etc.)
Rate of return of education
Since graduation from each level of education (except elementary) has a high return
the high rate of drop-outs in secondary and university education has a large social cost.
The large number of university graduates unemployed or underemployed imply an
important waste or public resources.
The poor have a lower chance to get the large return to higher education, conditional
on finding an adequate job.
Recommendations
When the unemployment rate is factor in the rate of return of higher education
graduates and high school graduates is not so different. Given the low level of
knowledge showed in international test, the expensive public higher education system
and the return of education of high school graduates the allocation of funds should grow
faster in secondary education than in higher education.
REFERENCES
Acebo, C. (2000), Technical background paper no 1 for the PESS, Statistical Annex.
ADB (2003), Key Indicators 2003: education for global participation.
Asian Development Bank and The World Bank (1998), Philippines education for the 21 st
century: the 1998 Philippines education sector study.
Blanchard, Olivier, and Lawrence F. Katz, 1992, "Regional Evolutions," Brookings
Papers on Economic Activity: 1, Brookings Institution, pp. 1-61.
Chapman, D. and D. Adams, Education in Developing Asia. The quality of education:
dimensions and strategies, ADB and University of Hong Kong.
Congressional Planning and Budget Committee (CPBC), House of Representatives of
Philippines (2003), Sectoral Budget Analysis (supplement to the Analysis of the
Presidents Budget for FY2003).
Cororatan, (2002), Research and development and technology in the Philippines, PIDS
working paper 2002-23.
Demery, L. (2000), Benefit incidence: a practitioners guide, Poverty and social
Development Group, Africa Region, The World Bank.
Esguerra, J., Balisican, A. and N. Confesor (2000), The Asian Crisis and te labor
market: Philippines case study, mimeo.
Government of Philippines (2000), Education for all: Philippines Assessment Report.
Gulosino, C. (2002), Evaluating private higher education in the Philippines: the case
for choice, equity and efficiency, Occasional Paper n. 68, National Center for the
Study of Privatization in Education, Columbia University.
Hanushek, E. and J. Luque (2001), Efficiency and equity in schools around the world,
mimeo, Stanford University.
Kraft, a. (2003), Determining the social rates of return to investment in basic social
services, mimeo.
Lpez, R., Thomas, V. and Y. Wang (1998), Addressing the education puzzle: the
distribution of education and economic reform, Policy Research Working Paper
2031, The World Bank.
Mauro, Paolo, and Spilimbergo Antonio, 1999, How Do the Skilled and Unskilled
Respond to Regional Shocks? The Case of Spain, IMF Staff Papers, Vol. 46, No. 1.
Morada, H. and T. Manzala (2001), Mismatches in the Philippines labor market,
mimeo, BLES.
Orbeta, A. (2002), Education, labor market and development: a review of the trends
and issues in the Philippines for the past 25 years, PIDS discussion paper 2002-19.
Presidential Commission on Education Reform (2000), Philippine Agenda for
Educational Reform: the PCER report.
Tan, E. (2001), The political economy of education reform, mimeo.
Tan, E. (2002), Studies in the access of poor to higher education, Background paper for
the Asian Development Bank.
Tan, E. (2003), School fee structure and inflation in Philippines higher education,
PIDS 2003-03.
UNESCO (2002), The EFA assessment country report: Philippines.
The World Bank (2001), Filipino report cards on pro-poor services.
The World Bank (2004), Implementing recent policy recommendations in education: a
review of progress, mimeo.
Technical note I.
Matching the APIS 2002 and the October LSF 2002
The matching is done on the basis of five basic variables: the household code (notice
you cannot use only this code since it is repeated in each region; for instance there is
one hcn=1021 in each region), region, province, barangany and line number (identifies
the individual line number, which is the number assign to the individual in the family.
The variables in the LFS are called hcn region prov bgy ln0.
Notice however that the number of observations is not exactly the same: the LFS has
196,482 observations while the APIS 2002 has 190,497.
There are also some inconsistencies related with the family size. For instance the first
observations of the LFS 2002 sorted by the variables above are (in bold missing in the
APIS 2002):
+----------------------------------+
| hcn region prov bgy ln0 |
|----------------------------------|
1. | 1 1 28 0381 1 |
2. | 1 1 28 0381 2 |
3. | 1 2 9 0011 1 |
4. | 1 3 8 0252 1 |
5. | 1 3 8 0252 2 |
|----------------------------------|
6. | 1 3 8 0252 3 |
7. | 1 3 8 0252 4 |
8. | 1 3 8 0252 5 |
9. | 1 3 8 0252 6 |
10. | 1 4 10 0041 1 |
|----------------------------------|
11. | 1 4 10 0041 2 |
12. | 1 4 10 0041 3 |
13. | 1 5 5 0061 1 |
14. | 1 5 5 0061 2 |
15. | 1 5 5 0061 3 |
|----------------------------------|
16. | 1 5 5 0061 4 |
17. | 1 5 5 0061 5 |
18. | 1 6 4 0111 1 |
19. | 1 6 4 0111 2 |
20. | 1 7 12 0451 1 |
|----------------------------------|
21. | 1 7 12 0451 2 |
22. | 1 7 12 0451 3 |
23. | 1 7 12 0451 4 |
24. | 1 7 12 0451 5 |
25. | 1 7 12 0451 6 |
|----------------------------------|
26. | 1 8 26 0221 1 |
27. | 1 8 26 0221 2 |
28. | 1 8 26 0221 3 |
29. | 1 8 26 0221 4 |
30. | 1 8 26 0221 5 |
|----------------------------------|
31. | 1 8 26 0221 6 |
32. | 1 9 7 0051 1 |
33. | 1 9 7 0051 2 |
34. | 1 9 7 0051 3 |
35. | 1 9 7 0051 4 |
|----------------------------------|
36. | 1 9 7 0051 5 |
37. | 1 10 13 0231 1 |
38. | 1 10 13 0231 2 |
39. | 1 10 13 0231 3 |
40. | 1 10 13 0231 4 |
|----------------------------------|
41. | 1 11 23 0114 1 |
42. | 1 11 23 0114 2 |
43. | 1 11 23 0114 3 |
44. | 1 11 23 0114 4 |
45. | 1 11 23 0114 5 |
|----------------------------------|
46. | 1 11 23 0114 6 |
47. | 1 12 35 0154 1 |
48. | 1 12 35 0154 2 |
49. | 1 12 35 0154 3 |
50. | 1 12 35 0154 4 |
|----------------------------------|
51. | 1 12 35 0154 5 |
52. | 1 12 35 0154 6 |
53. | 1 12 35 0154 7 |
54. | 1 12 35 0154 8 |
55. | 1 13 39 0293 1 |
|----------------------------------|
56. | 1 13 39 0293 2 |
57. | 1 14 1 0251 1 |
58. | 1 15 36 0691 1 |
59. | 1 15 36 0691 2 |
60. | 1 15 36 0691 3 |
|----------------------------------|
61. | 1 15 36 0691 4 |
62. | 1 15 36 0691 5 |
63. | 1 16 2 0151 1 |
64. | 1 16 2 0151 2 |
65. | 1 16 2 0151 3 |
|----------------------------------|
66. | 1 16 2 0151 4 |
67. | 1 16 2 0151 5 |
68. | 1 16 2 0151 6 |
69. | 1 16 2 0151 7 |
70. | 1 16 2 0151 8 |
|----------------------------------|
71. | 1 16 2 0151 9 |
72. | 2 1 28 0381 1 |
73. | 2 1 28 0381 2 |
74. | 2 1 28 0381 3 |
75. | 2 1 28 0381 4 |
|----------------------------------|
76. | 2 1 28 0381 5 |
77. | 2 1 28 0381 6 |
78. | 2 2 9 0011 1 |
79. | 2 2 9 0011 2 |
80. | 2 2 9 0011 3 |
|----------------------------------|
81. | 2 2 9 0011 4 |
82. | 2 3 8 0252 1 |
83. | 2 3 8 0252 2 |
84. | 2 3 8 0252 3 |
85. | 2 3 8 0252 4 |
|----------------------------------|
86. | 2 4 10 0041 1 |
87. | 2 4 10 0041 2 |
88. | 2 4 10 0041 3 |
89. | 2 4 10 0041 4 |
90. | 2 4 10 0041 5 |
|----------------------------------|
91. | 2 4 10 0041 6 |
92. | 2 6 4 0111 1 |
93. | 2 6 4 0111 2 |
94. | 2 6 4 0111 3 |
95. | 2 7 12 0451 1 |
|----------------------------------|
96. | 2 7 12 0451 2 |
97. | 2 7 12 0451 3 |
98. | 2 8 26 0221 1 |
99. | 2 8 26 0221 2 |
100. | 2 8 26 0221 3 |
+----------------------------------+
.
end of do-file
The same 100 observations for the APIS 2002 are (in bold missing in LFS 2002)
+---------------------------------------------------------+
| hcn reg prov bgy lno |
|---------------------------------------------------------|
1. | 1 Ilocos Region Ilocos Norte 0381 1 |
2. | 1 Ilocos Region Ilocos Norte 0381 2 |
3. | 1 Cagayan Valley Batanes 0011 1 |
4. | 1 Central Luzon Bataan 0252 1 |
5. | 1 Central Luzon Bataan 0252 2 |
|---------------------------------------------------------|
6. | 1 Central Luzon Bataan 0252 3 |
7. | 1 Central Luzon Bataan 0252 4 |
8. | 1 Central Luzon Bataan 0252 5 |
9. | 1 Southern Tagalog Batangas 0041 1 |
10. | 1 Southern Tagalog Batangas 0041 2 |
|---------------------------------------------------------|
11. | 1 Southern Tagalog Batangas 0041 3 |
12. | 1 Bicol Region Albay 0061 1 |
13. | 1 Bicol Region Albay 0061 2 |
14. | 1 Bicol Region Albay 0061 3 |
.
end of do-file
PROPOSALS FOR ADDITIONAL ANALYSIS